Science

Prior to 2020:

Many studies had been done on mask efficacy for respiratory viruses prior to 2020. Most involved surgical/medical masks or (N95 type) respirators in healthcare settings and most considered the ability of these masks/respirators to protect the wearer. There had also been several studies done in community settings (typically within households, or in group living facilities such as college/university dormitories) and some of these studies considered the impact of masks as source control (i.e. to protect those around the wearer). Many of these studies looked at other interventions (such as hand washing/sanitising) in addition to masks and unfortunately, some did not consider the interventions separately. Note that the community studies were reviewed in a collaborative effort by the CDC and WHO early in 2020, and the conclusion was that masks were not effective as protection for the wearer or as source control (see the first link under “Studies Completed in 2020” below).

Assessing the Evidence
It is generally accepted that the best type of study to evaluate a medical device or intervention is a randomized controlled trial (RCT) in which participants are randomly assigned to one of two groups – one group has the intervention (in this case wears the mask) and the other group does not. In order to provide strong evidence, large review or meta-analysis studies that consider or analyse the results from many (ideally RCT) studies are used.

One such review/meta-analysis study was the Cochrane Review (originally published in 2007 but updated a few times, with the latest update prior to 2020 published in July 2011), which included 6 RCTs involving masks, none of which found an effect of masks alone (two found an effect of masks combined with hand hygiene for a certain sub-portion of the study). The review also included 7 case control studies involving masks, however according to the authors, only 1 of these was considered to be at low risk of bias (4 were considered medium risk and 2 were considered high risk) and although the authors of that study showed a protective effect of masks, the analysis in the review on the data from that study suggested that there was not a significant effect.

Another systematic review study (bin-Reza et al.) undertaken to inform pandemic influenza guidance in the UK and published in December 2011, concluded that “there is a limited evidence base to support the use of masks and/or respirators in healthcare or community settings”.

Another way to gauge the level of evidence when there are many diverse studies with differing interventions, measures, and outcomes, is to look at decisions within the justice system that are made based on the available evidence and expert testimony regarding such evidence. In Canada, in 2012 there was a push by Health Canada to require nurses who chose not to get an influenza vaccine to wear surgical masks throughout their shifts during the influenza season and hospitals/healthcare organizations began to implement these ‘Vaccinate or Mask’ (VOM) policies. In two cases (one decided in 2015 and one in 2018), two separate Arbitrators examined the science and listened to expert testimony and each decided for the Ontario Nurses Association against such mandatory mask policies implemented by the employers. In the first (Hayes 2015), Arbitrator Hayes decided that “the weight of scientific evidence said to support the VOM Policy on patient safety grounds is insufficient to warrant the imposition of a mask-wearing requirement for up to six months every year.” In the second (Kaplan 2018), Arbitrator Kaplan described the evidence put forth in support of masking by the employer (in the report for and testimony during) the proceeding as “insufficient, inadequate, and completely unpersuasive”. He further concluded that “the preponderance of the masking evidence is compelling – surgical and procedural masks are extremely limited in terms of source control: they do not prevent the transmission of the influenza virus.”

Studies of particular interest:
None of the RCT studies considering masks as source control found a significant impact of masks alone. Although some studies (for example Suess et al., Cowling et al.) found a significant protective effect of masks combined with ‘hand hygiene’ in certain posthoc analyses (for example by controlling for certain variables, combining groups, and/or restricting the analysis to those participants that indicated that they began the intervention/s within a certain time frame), a closer look at one of these studies (Cowling et al.) suggests that masks did not contribute to the decrease in infection and in fact may have actually had a detrimental effect. This can be seen by comparing the results of the ‘facemasks plus hand hygiene’ group to the ‘hand hygiene only’ group in the data tables where it is evident that masks may have actually reduced the beneficial effect of hand hygiene on respiratory infection rates.
Similarly, the only RCT involving cloth masks (MacIntyre et al.) suggests that cloth masks may have a detrimental effect and concludes that “the results caution against the use of cloth masks”.
Although not an RCT, this lab study (Leung et al.) is particularly relevant because it examines individuals infected with various respiratory viruses – seasonal coronaviruses (the common cold, not Covid-19), influenza viruses, and rhinoviruses and measures the viral particles emitted in their exhaled breath during a 30-minute collection period. The interesting finding is that for participants infected with coronaviruses only those who coughed during their breath collection emitted any viral particles and for participants infected with influenza only one of nine that did not cough during their breath collection emitted any viral particles. Also, for participants infected with rhinoviruses, the authors found “no significant differences between detection of virus with or without face masks”. They also found that the majority of coronavirus and influenza patients (including those who coughed during the collection) did not emit any viral particles in their 30-minute breath collection and those (participants with all three types of respiratory virus infections) who did emit viral particles, showed a low viral load. They concluded that these results “might imply that prolonged close contact would be required for transmission to occur, even if transmission was primarily via aerosols, as has been described for rhinovirus colds”. So what this study tells us is that for respiratory viruses that are spread through droplets, even symptomatic patients do not emit viral particles when they are not coughing or sneezing, so masks will be of no benefit for asymptomatic patients (or for symptomatic patients that are following proper etiquette when they cough and/or sneeze). And for viruses that are spread through aerosols, a surgical mask will not be of benefit because it can not block the aerosols due to their size. In either case, masks are of no use.

Studies completed in 2020:

Review or Meta-Analysis Studies
Early in 2020 the CDC, in collaboration with the WHO, conducted a review and meta-analysis (Xiao et al.) on the effectiveness of non-pharmaceutical protective measures against pandemic influenza in non-healthcare settings, intended to shape the guidelines of the WHO “on the use of nonpharmaceutical interventions for pandemic influenza in non-medical settings”. The review authors concluded that they “did not find evidence that surgical-type face masks are effective in reducing laboratory-confirmed influenza transmission, either when worn by infected persons (source control) or by persons in the general community to reduce their susceptibility”.
Later in 2020, the Cochrane Review mentioned earlier was updated (Jefferson et al.) to include an additional 44 new RCTs and cluster‐RCTs and the findings were clear – in terms of surgical/medical masks, “wearing a mask may make little to no difference in how many people caught a flu‐like illness (9 studies; 3507 people); and probably makes no difference in how many people have flu confirmed by a laboratory test (6 studies; 3005 people). “

Individual Studies of Particular Interest
An RCT on the use of medical masks in the community to protect against Covid-19 (Bundgaard et al.), found that masks made no significant impact against the virus when added to other measures.